NeighborMom() vs AnyMom()

The way a mother is chosen after death has impacts on the result of the simulation. Below are two methods for selecting that individual to reproduce: weighted sampling by fitness of all individuals (anyMom), or weighted sampling by fitness of only individuals adjacent to the dead one (neighborMom).

function neighborMom(network::NetworkParameters, kID::Int64)
    fitWeights = StatsBase.weights([network.popFitness[kID%100+1], network.popFitness[(kID+98)%100+1]]) #index plus/minus one, loops around 100
    momIndex = (kID + sample([1, -1], fitWeights) + 99)%100 + 1 #have to use (idx +99)%100 + 1 to loop from 100 to 1 properly
    momIndex
end

function anyMom(network::NetworkParameters, kID::Int64)
    network.popFitness[kID] = 0
    fitWeights = StatsBase.weights(network.popFitness)
    momIndex = sample(1:network.popSize, fitWeights)
    resolveLocs(network, kID, momIndex) #updates locations, ensures each individual has a unique location by inserting the child adjacent to the parent and pushing others away 
    momIndex
end

UndefVarError: NetworkParameters not defined



Stacktrace:

 [1] top-level scope

   @ c:\Users\jonah\OneDrive\Desktop\Research\Parker Model\neighbor-mom.ipynb:1

Both NeighborMom() and AnyMom() serve as different methods replacing FindMom() based on our method of selecting parents. NeighborMom() only compares fitnesses of neighboring individuals, while AnyMom() looks at fitnesses of all individuals in the population. Locations do not need to be separately updated for NeighborMom(), as each individual is automatically adjacent to a parent.

Old heatmap for Iterating by Ben/CL

using JLD2
using StatsBase
using Plots
#backend(:plotly)
using DataFrames
using CSV

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr) 
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 1 method)

High Strength of Selection, AnyMom(), no Distance-based inheritance

The following two runs were created before distance inheritance was implemented, so do not have the mean connection distance and have to use an older hmap() function.

file = "anyMom_no_pnd_50_reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 1.0), 
hmap(file, 20, 8, "PR", 0.0, 0.3),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 1 High strength of selection with the mother selected from any individual, pnd and prd not evolving separately. The results above were created with: C=0.5, D=0.0, gen=100000, pn=0.5, distInherit=false, pnd=false, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5.

file = "anyMom_pnd_50_reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.3),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 2 High strength of selection with the mother selected from any individual, pnd but not prd evolving separately. The results above were created with: C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, findMom=anyMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5.

New Heatmap Function

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 1 method)

High Strength of Selection, NeighborMom(), no Distance-based inheritance

file = "neighbor_mom_no_pnd_fixed.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 0.7), 
hmap(file, 20, 8, "PR", 0.0, 0.40),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.25), 
hmap(file, 20, 11, "Connected Components", 1.0, 35.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)
]
#,

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 3 High strength of selection with the mother selected from dead individuals' neighbors, pnd and prd not evolving separately. The results above were created with: C=0.5, D=0.0, gen=100000, distInherit=false, distFactor=1.0, pn=0.5, findMom=neighborMom, pnd=false, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10). Initial cooperation frequency was 0.5 .

file = "neighbor_mom_pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.50),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.25), 
hmap(file, 20, 11, "Connected Components", 1.0, 30.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 100.0)
]
#,

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 4 High strength of selection with the mother selected from dead individuals' neighbors, pnd but not prd evolving separately. The results above were created with: C=0.5, D=0.0, gen=100000, distInherit=false, distFactor=1.0, pn=0.5, findMom=neighborMom, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5 .

Difference Between NeighborMom and anyMom

Heatmap for neighborMom - anyMom

function benClDataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    dataArr
end

function benclDataReadMinus(data, data2, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    #print(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    benclData = benClDataRead(data2, range, index)
    dataArr -= benclData
    dataArr
	#return transpose(dataArr[:, :, 1]) #ensures axes are correct
end

function hmap(data::String, data2::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = benclDataReadMinus(data, data2, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, c=cgrad(:bluesreds), xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

hmap (generic function with 2 methods)

newdata = "neighbor_mom_no_pnd_fixed.csv"
compData = "anyMom_no_pnd_new.csv"

fig1hmaps = [hmap(newdata, compData, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 20, 6, "Δ PN", -0.6, 0.6), 
hmap(newdata, compData, 20, 8, "Δ PR", -0.35, 0.35),
hmap(newdata, compData, 20, 10, "Δ Mean Shortest Path", -2.0, 2.0), 
hmap(newdata, compData, 20, 11, "Δ Connected Components", -35.0, 35.0), 
hmap(newdata, compData, 20, 12, "Δ Mean Conn. Comp. Size", -65.0, 65.0), 
hmap(newdata, compData, 20, 13, "Δ Largest Conn. Comp.", -45.0, 45.0),
hmap(newdata, compData, 20, 4, "Δ Degree", -30.0, 30.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 5 Difference between Figure 3 and Figure 1 (parents selected from neighbors versus entire population, PND not evolving separately from PNC)

newdata = "neighbor_mom_pnd.csv"
compData = "anyMom_pnd_new.csv"

fig1hmaps = [hmap(newdata, compData, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 20, 6, "Δ PNC", -0.4, 0.4), 
hmap(newdata, compData, 20, 7, "Δ PND", -0.4, 0.4),
hmap(newdata, compData, 20, 8, "Δ PR", -0.1, 0.1),
hmap(newdata, compData, 20, 10, "Δ Mean Shortest Path", -2.0, 2.0), 
hmap(newdata, compData, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 20, 12, "Δ Mean Conn. Comp. Size", -65.0, 65.0), 
hmap(newdata, compData, 20, 13, "Δ Largest Conn. Comp.", -45.0, 45.0),
hmap(newdata, compData, 20, 4, "Δ Degree", -30.0, 30.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 6 Difference between Figure 4 and Figure 2 (parents selected from neighbors versus entire population, PND evolving separately from PNC)

NeighborRange

The following introduces a new parameter, neighborRange, and modifies neighborMom() as below to allow for an input range to control how many individuals are possible to be selected for the mother.

function neighborMom(network::NetworkParameters, kID::Int64, neighborRange::Int64=1)
    indexes = zeros(Int64, 2*neighborRange)
    fitnesses = zeros(2*neighborRange)
    for(i) in 1:neighborRange
        indexes[i] = (kID-i+99)%100+1 #index minus 1 through neighborRange, loops around 100
        indexes[i+neighborRange] = (kID+i+99)%100+1 #index plus 1 through neighborRange, loops around 100
        fitnesses[i] = network.popFitness[indexes[i]] 
        fitnesses[i+neighborRange] = network.popFitness[indexes[i+neighborRange]]
    end
    fitWeights = StatsBase.weights(fitnesses)
    return sample(indexes, fitWeights) 
end

UndefVarError: NetworkParameters not defined



Stacktrace:

 [1] top-level scope

   @ c:\Users\jonah\OneDrive\Desktop\Research\Parker Model\neighbor-mom.ipynb:1

Heatmap for iterating by cl/neighborRange (1-50)

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    sort!(df, [:distNeighborRange, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range 
        push!(y_axis, (string(Int(floor(-1+51*i/range)))))
        push!(x_axis, (string(round(0.4*i/range; digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="Neighbor Range", clim=(cLow,chI))
    p
end

hmap (generic function with 2 methods)

Low benefit

file = "NR-50-no-pnd-more-reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 0.6), 
hmap(file, 20, 8, "PR", 0.0, 0.3),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.5), 
hmap(file, 20, 11, "Connected Components", 1.0, 40.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 27.5)
]
#hmap(file, 20, 4, "Degree", 1.0, 35.0)
plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 7 High strength of selection with the mother selected from individuals a distance neighborRange or below from, neither pnd nor prd evolving separately. The results above were created with: B=2.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, distFactor=1.0, findMom=neighborMom, pn=0.5, pnd=false, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5.

The exact values of the y-axis are: 1 4 6 9 11 14 16 19 21 24 27 29 32 34 37 39 42 44 47 50

file = "neighborRange-1-50-more-reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 0.8), 
hmap(file, 20, 7, "PND", 0.0, 0.8),
hmap(file, 20, 8, "PR", 0.0, 0.6),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 6.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 40.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 35.0)
]
#
plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 8 High strength of selection with the mother selected from individuals a distance neighborRange or below from, pnd but not prd evolving separately. The results above were created with: B=2.0, C=0.5, D=0.0, gen=100000, distInherit=false, distFactor=1.0, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5.

The exact values of the y-axis are: 1 4 6 9 11 14 16 19 21 24 27 29 32 34 37 39 42 44 47 50

Finding the difference from both runs with anyMom() and neighborRange from 1-50. New heatmap below.

High benefit

file = "NR-50-9.0-ben-no-pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 0.8), 
hmap(file, 20, 8, "PR", 0.0, 0.35),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 2.5), 
hmap(file, 20, 11, "Connected Components", 1.0, 2.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 62.5)
]
#hmap(file, 20, 4, "Degree", 1.0, 35.0)
plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 9 High strength of selection with the mother selected from individuals a distance neighborRange or below from, with neither pnd nor prd evolving separately. The results above were created with: B=9.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, distFactor=1.0, findMom=neighborMom, pn=0.5, pnd=false, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5 Initial cooperation frequency was 0.5

The exact values of the y-axis are: 1 4 6 9 11 14 16 19 21 24 27 29 32 34 37 39 42 44 47 50

file = "NR-50-9.0-ben-pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.5),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 2.5), 
hmap(file, 20, 11, "Connected Components", 1.0, 2.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 1.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 62.5)
]
#
plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 10 High strength of selection with the mother selected from individuals a distance neighborRange or below from, pnd but not prd evolving separately. The results above were created with: B=9.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, distFactor=1.0, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5

The exact values of the y-axis are: 1 4 6 9 11 14 16 19 21 24 27 29 32 34 37 39 42 44 47 50

NeighborRange Differences

Heatmap for the difference between NeighborRange plots and runs with anyMom

function benClDataRead2(data, data2row, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    #df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    df = df[!, [3, 2, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        dataArr[i, 1, 1] = dfData[i+range*data2row, index]
    end
    #print(dataArr)
    dataArr
end

function dataReadMinus(data, data2, data2row, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    sort!(df, [:cl, :distNeighborRange])
    dfData = Matrix{Float64}(df)
    #print(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    benclData = benClDataRead2(data2, data2row, range, index)
    for(i) in 1:range
        #println(benclData[i, 4, 1], " subt from ", dataArr[i, 20, 1], " equals ", dataArr[i, 20, 1]-benclData[i, 4, 1])
        dataArr[i, :, 1] .-= benclData[i, 1, 1]
    end   
	tempData = transpose(dataArr[:, :, 1])
    return tempData
end

function hmap(data::String, data2::String, data2row::Int, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataReadMinus(data, data2, data2row, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range 
        push!(y_axis, (string(Int(floor(-1+51*i/range)))))
        push!(x_axis, (string(round(0.4*i/range; digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :]; title = titles, c=cgrad(:bluesreds), xlabel="CL",ylabel="Neighbor Range", clim=(cLow,chI))
    p
end

hmap (generic function with 3 methods)

Low Benefit

newdata = "NR-50-no-pnd-more-reps.csv"
compData = "anyMom_no_pnd_50_reps.csv"

fig1hmaps = [hmap(newdata, compData, 4, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 6, "Δ PN", -0.25, 0.25), 
hmap(newdata, compData, 4, 20, 8, "Δ PR", -0.06, 0.06),
hmap(newdata, compData, 4, 20, 10, "Δ Mean Shortest Path", -2.0, 2.0), 
hmap(newdata, compData, 4, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 4, 20, 12, "Δ Mean Conn. Comp. Size", -50.0, 50.0), 
hmap(newdata, compData, 4, 20, 13, "Δ Largest Conn. Comp.", -45.0, 45.0),
hmap(newdata, compData, 4, 20, 4, "Δ Degree", -5.0, 5.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 11 Difference between Figure 7 (with neighborRange) and Figure 1 (anyMom), both without PND evolving separately and benefit=2.0

newdata = "neighborRange-1-50-more-reps.csv"
compData = "anyMom_pnd_50_reps.csv"

fig1hmaps = [hmap(newdata, compData, 4, 20, 3, "Δ Cooperation Frequency", -0.25, 0.25), 
hmap(newdata, compData, 4, 20, 6, "Δ PNC", -0.3, 0.3), 
hmap(newdata, compData, 4, 20, 7, "Δ PND", -0.3, 0.3),
hmap(newdata, compData, 4, 20, 8, "Δ PR", -0.1, 0.1),
hmap(newdata, compData, 4, 20, 10, "Δ Mean Shortest Path", -2.0, 2.0), 
hmap(newdata, compData, 4, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 4, 20, 12, "Δ Mean Conn. Comp. Size", -40.0, 40.0), 
hmap(newdata, compData, 4, 20, 13, "Δ Largest Conn. Comp.", -35.0, 35.0),
hmap(newdata, compData, 4, 20, 4, "Δ Degree", -5.0, 5.0)
#hmap(newdata, compData, 20, 5, "Δ Fitness", -100.0, 100.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 12 Difference between Figure 8 (with neighborRange) and Figure 2 (anyMom), both with PND evolving separately and benefit=2.0

High Benefit Differences

newdata = "NR-50-9.0-ben-no-pnd.csv"
compData = "anyMom_no_pnd_50_reps.csv"

fig1hmaps = [hmap(newdata, compData, 18, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 6, "Δ PN", -0.45, 0.45), 
hmap(newdata, compData, 18, 20, 8, "Δ PR", -0.2, 0.2),
hmap(newdata, compData, 18, 20, 10, "Δ Mean Shortest Path", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 11, "Δ Connected Components", -4.0, 4.0), 
hmap(newdata, compData, 18, 20, 12, "Δ Mean Conn. Comp. Size", -20.0, 20.0), 
hmap(newdata, compData, 18, 20, 13, "Δ Largest Conn. Comp.", -15.0, 15.0),
hmap(newdata, compData, 18, 20, 4, "Δ Degree", -15.0, 15.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 13 Difference between Figure 9 (with neighborRange) and Figure 1 (anyMom), both without PND evolving separately and benefit=9.0

newdata = "NR-50-9.0-ben-pnd.csv"
compData = "anyMom_pnd_50_reps.csv"

fig1hmaps = [hmap(newdata, compData, 18, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 6, "Δ PNC", -0.45, 0.45), 
hmap(newdata, compData, 18, 20, 6, "Δ PND", -0.45, 0.45), 
hmap(newdata, compData, 18, 20, 8, "Δ PR", -0.2, 0.2),
hmap(newdata, compData, 18, 20, 10, "Δ Mean Shortest Path", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 11, "Δ Connected Components", -4.0, 4.0), 
hmap(newdata, compData, 18, 20, 12, "Δ Mean Conn. Comp. Size", -20.0, 20.0), 
hmap(newdata, compData, 18, 20, 13, "Δ Largest Conn. Comp.", -15.0, 15.0),
hmap(newdata, compData, 18, 20, 4, "Δ Degree", -15.0, 15.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 14 Difference between Figure 10 (with neighborRange) and Figure 2 (anyMom), both without PND evolving separately and benefit=9.0

Distance-Based Inheritance

The next section involves physical distance having an impact on forming connections. As distance increases, individuals are exponentially less likely to be connected to by a newborn.

Probability to connect = distFactor^n * P(N/R)(C/D), where n is the shortest distance along a torus between individuals

Heatmap for iterating by distFactor/cl

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    sort!(df, [:distFactor, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range 
        push!(y_axis, (string(round(0.75+0.25*i/range; digits = 3)))) 
        push!(x_axis, (string(round(0.4*i/range; digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="DistFactor", clim=(cLow,chI))
    p
end

hmap (generic function with 3 methods)

Low Benefit

file =  "distInherit_nmon_no_pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 0.7), 
hmap(file, 20, 8, "PR", 0.0, 0.4),
hmap(file, 20, 14, "Mean Dist. of Connections", 2.5, 27.5),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 5.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 40.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 30.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 15 High strength of selection with the probability of forming connections, with neither pnd nor prd evolving separately, and at low benefit. The results above were created with: B=2.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=true, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5.

file =  "distInherit_nmon_pnd_50_reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 0.7),
hmap(file, 20, 7, "PND", 0.0, 0.7), 
hmap(file, 20, 8, "PR", 0.0, 0.4),
hmap(file, 20, 14, "Mean Dist. of Connections", 2.5, 27.5),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 5.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 40.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 30.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 16 High strength of selection with the probability of forming connections, pnd but not prd evolving separately, and at low benefit. The results above were created with: B=2.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=true, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5.

High Benefit

file = "distInherit_nmon_9.0_ben_pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PN", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.8),
hmap(file, 20, 14, "Mean Dist. of Connections", 2.5, 27.5),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 5.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 5.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 30.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 17 High strength of selection with the probability of forming connections, with neither pnd nor prd evolving separately, and at high benefit. The results above were created with: B=9.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=true, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5.

file =  "distInherit_nmon_9.0_ben_no_pnd.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0),
hmap(file, 20, 7, "PND", 0.0, 1.0), 
hmap(file, 20, 8, "PR", 0.0, 0.8),
hmap(file, 20, 14, "Mean Dist. of Connections", 2.5, 27.5),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 5.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 5.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 30.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 18 High strength of selection with the probability of forming connections, pnd but not prd evolving separately, and at high benefit. The results above were created with: B=9.0, C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=true, findMom=neighborMom, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5

Differences From Distance-based Inheritance

Difference heatmap for iterating by distFactor/cl

function benClDataRead2(data, data2row, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    #df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    df = df[!, [3, 2, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        dataArr[i, 1, 1] = dfData[i+range*data2row, index]
    end
    #print(dataArr)
    dataArr
end

function dataReadMinus(data, data2, data2row, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 6, 4, 5, 8, 13, 14, 15, 16, 17, 3, 11, 10, 12, 7, 9, 1]] #used for soring by clratio/distFactor/distNeighborRange, including graphCalcs
    sort!(df, [:cl, :distFactor])
    dfData = Matrix{Float64}(df)
    #print(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    benclData = benClDataRead2(data2, data2row, range, index)
    for(i) in 1:range
        #println(benclData[i, 4, 1], " subt from ", dataArr[i, 20, 1], " equals ", dataArr[i, 20, 1]-benclData[i, 4, 1])
        dataArr[i, :, 1] .-= benclData[i, 1, 1]
    end   
	tempData = transpose(dataArr[:, :, 1])
    return tempData
end

function hmap(data::String, data2::String, data2row::Int, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataReadMinus(data, data2, data2row, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range 
        push!(y_axis, (string(round(0.75+0.25*i/range; digits = 3)))) 
        push!(x_axis, (string(round(0.4*i/range; digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :]; title = titles, c=cgrad(:bluesreds), xlabel="CL",ylabel="DistFactor", clim=(cLow,chI))
    p
end

hmap (generic function with 3 methods)

newdata = "distInherit_nmon_no_pnd.csv"
compData = "neighbor_mom_no_pnd_fixed.csv"

fig1hmaps = [hmap(newdata, compData, 4, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 6, "Δ PN", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 8, "Δ PR", -0.3, 0.3),
hmap(newdata, compData, 4, 20, 10, "Δ Mean Shortest Path", -4.0, 4.0), 
hmap(newdata, compData, 4, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 4, 20, 12, "Δ Mean Conn. Comp. Size", -50.0, 50.0), 
hmap(newdata, compData, 4, 20, 13, "Δ Largest Conn. Comp.", -50.0, 50.0),
hmap(newdata, compData, 4, 20, 4, "Δ Degree", -10.0, 10.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 19 Difference between Figure 15 (with distFactor) and Figure 3 (neighborMom), both without PND evolving separately and benefit=2.0

newdata = "distInherit_nmon_pnd_50_reps.csv"
compData = "neighbor_mom_pnd.csv"

fig1hmaps = [hmap(newdata, compData, 4, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 6, "Δ PNC", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 7, "Δ PND", -0.5, 0.5), 
hmap(newdata, compData, 4, 20, 8, "Δ PR", -0.3, 0.3),
hmap(newdata, compData, 4, 20, 10, "Δ Mean Shortest Path", -4.0, 4.0), 
hmap(newdata, compData, 4, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 4, 20, 12, "Δ Mean Conn. Comp. Size", -50.0, 50.0), 
hmap(newdata, compData, 4, 20, 13, "Δ Largest Conn. Comp.", -50.0, 50.0),
hmap(newdata, compData, 4, 20, 4, "Δ Degree", -10.0, 10.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 20 Difference between Figure 16 (with distFactor) and Figure 4 (neighborMom),, both with PND evolving separately from PNC and benefit=2.0

newData = "distInherit_nmon_9.0_ben_pnd.csv"
compData = "neighbor_mom_no_pnd_fixed.csv"

fig1hmaps = [hmap(newdata, compData, 18, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 6, "Δ PN", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 8, "Δ PR", -0.3, 0.3),
hmap(newdata, compData, 18, 20, 10, "Δ Mean Shortest Path", -4.0, 4.0), 
hmap(newdata, compData, 18, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 18, 20, 12, "Δ Mean Conn. Comp. Size", -50.0, 50.0), 
hmap(newdata, compData, 18, 20, 13, "Δ Largest Conn. Comp.", -50.0, 50.0),
hmap(newdata, compData, 18, 20, 4, "Δ Degree", -10.0, 10.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (4,2), size = (1000, 1200))

Figure 21 Difference between Figure 17 (with distFactor) and Figure 3 (neighborMom), both without PND evolving separately and benefit=2.0

newdata = "distInherit_nmon_9.0_ben_no_pnd.csv"
compData = "neighbor_mom_pnd.csv"

fig1hmaps = [hmap(newdata, compData, 18, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 6, "Δ PNC", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 7, "Δ PND", -0.5, 0.5), 
hmap(newdata, compData, 18, 20, 8, "Δ PR", -0.3, 0.3),
hmap(newdata, compData, 18, 20, 10, "Δ Mean Shortest Path", -4.0, 4.0), 
hmap(newdata, compData, 18, 20, 11, "Δ Connected Components", -25.0, 25.0), 
hmap(newdata, compData, 18, 20, 12, "Δ Mean Conn. Comp. Size", -50.0, 50.0), 
hmap(newdata, compData, 18, 20, 13, "Δ Largest Conn. Comp.", -50.0, 50.0),
hmap(newdata, compData, 18, 20, 4, "Δ Degree", -10.0, 10.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Figure 22 Difference between Figure 18 (with distFactor) and Figure 4 (neighborMom), both with PND evolving separately from PNC and benefit=9.0

Stabilization Testing

The following were testing iterating by number of generations (50000, 75000, 100000, 150000, 250000) and cl, to see when cooperation frequency stabilizes.

Heatmap for Iterating by Gen/CL

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    gens = [50000, 75000, 100000, 150000, 250000]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(gens[i])))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="Generations", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 3 methods)

file = "testing-anymom.csv"
fig1hmaps = [hmap(file, 5, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 5, 6, "PNC", 0.0, 0.8), 
hmap(file, 5, 7, "PND", 0.0, 0.8),
hmap(file, 5, 8, "PR", 0.0, 0.04),
hmap(file, 5, 11, "Connected Components", 1.0, 40.0), 
hmap(file, 5, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
]
#,

plot(fig1hmaps..., colorbar = true, grid=false, layout = (3,2), size = (1000, 900))

Supplementary Figure 1 High strength of selection with the mother selected from any individuals, pnd but not prd evolving separately. The results above were created with: B=2.0, C=0.5, D=0.0, CL=pard["cl"], gen=Int(pard["ben"]), findMom=anyMom, distInherit=false, distFactor=0.975, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5

file = "testing-nmom.csv"
fig1hmaps = [hmap(file, 5, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 5, 6, "PNC", 0.0, 0.9), 
hmap(file, 5, 7, "PND", 0.0, 0.9),
hmap(file, 5, 8, "PR", 0.0, 0.14),
hmap(file, 5, 11, "Connected Components", 1.0, 20.0), 
hmap(file, 5, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
]
#,

plot(fig1hmaps..., colorbar = true, grid=false, layout = (3,2), size = (1000, 900))

Supplementary Figure 2 High strength of selection with the mother selected from dead individuals' neighbors, pnd but not prd evolving separately. The results above were created with: B=2.0, C=0.5, D=0.0, CL=pard["cl"], gen=pard["ben"], findMom=neighborMom, distInherit=false, distFactor=0.975, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5 .

PRD Evolving Separately

While using anyMom and high strength of selection (conditions similar to Figures 1-2)

Heatmap for anyMom

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

hmap (generic function with 3 methods)

file = "anyMom_prd_50_reps.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 4, "Degree", 1.0, 75.0),
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0), 
hmap(file, 20, 8, "PRC", 0.0, 1.0),
hmap(file, 20, 9, "PRD", 0.0, 1.0),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Supplemental Figure 3 High strength of selection with the mother selected from any individual, pnd and prd evolving separately. The results above were created with: C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, distFactor=1.0, findMom=anyMom, pn=0.5, pnd=false, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50. Initial cooperation frequency was 0.5. Initial cooperation frequency was 0.5.

1 dies, 10 selected as parent (with neighborRange=10) 95 dies, 10 can't be selected as parent but 1 can distInheritance resolve location - baby was set next to parent

Distance Between Individuals

Graphs of the mean distance between individuals along a torus, from selected runs above.

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end

function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 3 methods)

file = "fixed_connDist_test_full.csv"
file2 = "fixed_connDist_nmom_pnd.csv"
fig1hmaps = [
hmap(file, 20, 14, "Mean Dist. of Connections, AnyMom", 10.0, 26.0),
hmap(file, 20, 4, "Degree, AnyMom", 1.0, 75.0),
hmap(file2, 20, 14, "Mean Dist. of Connections, NeighborMom", 10.0, 26.0),
hmap(file2, 20, 4, "Degree, NeighborMom", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (2,2), size = (1000, 600))

Supplemental Figure 4 Mean distance of connections and degree for runs using anyMom and neighborMom. All other inputs were controlled at C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, distInherit=false, pn=0.5, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5.

Heatmap for Difference of Ben/CL

function benClDataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    dataArr
end

function benclDataReadMinus(data, data2, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [2, 3, 5, 6, 8, 13, 14, 15, 16, 17, 4, 11, 10, 12, 7, 9, 1]] #used for soring by ben/cl, including graphCalcs
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    #print(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    benclData = benClDataRead(data2, range, index)
    dataArr -= benclData
    dataArr
	#return transpose(dataArr[:, :, 1]) #ensures axes are correct
end

function hmap(data::String, data2::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 distance  16 inclusion  17 assortment 
    dataArr = benclDataReadMinus(data, data2, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, c=cgrad(:bluesreds), xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

hmap (generic function with 3 methods)

newdata = "nr50_pnd_whole_graph.csv"
compData = "anyMom_pnd_50_reps.csv"

fig1hmaps = [hmap(newdata, compData, 20, 3, "Δ Cooperation Frequency", -0.5, 0.5), 
hmap(newdata, compData, 20, 6, "Δ PNC", -0.6, 0.6), 
hmap(newdata, compData, 20, 7, "Δ PND", -0.6, 0.6), 
hmap(newdata, compData, 20, 8, "Δ PR", -0.35, 0.35),
hmap(newdata, compData, 20, 10, "Δ Mean Shortest Path", -2.0, 2.0), 
hmap(newdata, compData, 20, 11, "Δ Connected Components", -35.0, 35.0), 
hmap(newdata, compData, 20, 12, "Δ Mean Conn. Comp. Size", -65.0, 65.0), 
hmap(newdata, compData, 20, 13, "Δ Largest Conn. Comp.", -45.0, 45.0),
hmap(newdata, compData, 20, 4, "Δ Degree", -30.0, 30.0)
]

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Supplemental Figure 5 Difference between a run with neighborRange=50 and anyMom, with PND evolving separately. Constant inputs were C=0.5, D=0.0, gen=100000, pn=0.5, distInherit=false, pnd=true, pr=0.0001, prd=false, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=50

Allow Rejection of Connections

Changed Order With Allowing PNA/PRA

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [1, 2, 4, 5, 6, 12, 13, 15, 16, 17, 3, 9, 8, 10, 11, 14, 7]] #used for soring by ben/cl, including graphCalcs
    #print(df)
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end
pars = Dict([
            "ben" => 0.0,
            "cl" => 0.0,
            "pnc_end" => 0.0,
            "pnd_end" => 0.0,
            "pna_end" => 0.0,
            "prc_end" => 0.0,
            "prd_end" => 0.0,
            "pra_end" => 0.0,
            "degree" => 0.0,
            "inclusion" => 0.0,
            "coopFreq" => 0.0,
            "fitness" => 0.0,
            "shortestPath" => 0.0,
            "connComponents" => 0.0,
            "meanConnCompSize" => 0.0,
            "largestConnComp" => 0.0,
            "meanConnDistance" => 0.0,
        ])
function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 pna_end  16 pra_end  17 inclusion 
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 3 methods)

file = "pacc_pnd_checking.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.3),
hmap(file, 20, 15, "PNA", 0.0, 1.0),
hmap(file, 20, 16, "PRA", 0.0, 1.0),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (6,2), size = (1000, 1800))

Supplemental Figure 6 Run with PND evolving separately, after allowReject was implemented but without it enabled. Inputs were C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, findMom=neighborMom, neighborRange=50, pn=0.5, distInherit=false, pnd=true, pr=0.0001, prd=false, allowReject=false, pa=0.75, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5

file = "pacc_pnd_full.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.3),
hmap(file, 20, 15, "PNA", 0.0, 1.0),
hmap(file, 20, 16, "PRA", 0.0, 1.0),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (6,2), size = (1000, 1800))

Supplemental Figure 7 Run with PND evolving separately and allowReject enabled. Inputs were C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, findMom=neighborMom, neighborRange=50, pn=0.5, distInherit=false, pnd=true, pr=0.0001, prd=false, allowReject=true, pa=0.75, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5

file = "pacc_prd_full.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PRC", 0.0, 0.8),
hmap(file, 20, 8, "PRD", 0.0, 0.8),
hmap(file, 20, 15, "PNA", 0.0, 1.0),
hmap(file, 20, 16, "PRA", 0.0, 1.0),
hmap(file, 20, 10, "Mean Shortest Path", 1.0, 4.0), 
hmap(file, 20, 11, "Connected Components", 1.0, 45.0), 
hmap(file, 20, 12, "Mean Conn. Comp. Size", 1.0, 100.0), 
hmap(file, 20, 13, "Largest Conn. Comp.", 10.0, 100.0),
hmap(file, 20, 4, "Degree", 1.0, 75.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (6,2), size = (1000, 1800))

Supplemental Figure 7 Run with PRD and PND evolving separately from PRC and PNC and allowReject enabled. Inputs were C=0.5, D=0.0, gen=100000, dbOrder=deathbirth, findMom=neighborMom, neighborRange=50, pn=0.5, distInherit=false, pnd=true, pr=0.0001, prd=false, allowReject=true, pa=0.75, muP=0.01, delta=0.5, sigmapn=0.01, sigmapr=0.01, reps=10. Initial cooperation frequency was 0.5

Separate P(N/R)(C/D)'s

function dataRead(data, range, index::Int)
    df = CSV.read(data, DataFrame, delim=",")
    df = df[!, [1, 2, 4, 5, 6, 12, 13, 15, 16, 29, 3, 9, 8, 10, 11, 14, 7, 18, 19, 17, 21, 22, 20, 24, 25, 23, 27, 28, 26]] #used for soring by ben/cl, including graphCalcs
    #print(df)
    #df = df[!, [6, 9, 2, 3, 7, 8, 10, 4, 5, 1]] #used for when sorting by pn/r
    #sort!(df, [:pn, :pr]) #used for when sorting by pn/r
    sort!(df, [:ben, :cl])
    dfData = Matrix{Float64}(df)
    dataArr = zeros(range, range, 1) #transforms [range^2 * 10] arr to [range, range, 1] arr
    for(i) in 1:range
        for(j) in 1:range
            dataArr[i, j, 1] = dfData[range*(i-1)+j, index]
        end
    end
    #print(dataArr)
    dataArr
end
pars = Dict([
            "ben" => 0.0,
            "cl" => 0.0,
            "pnc_end" => 0.0,
            "pnd_end" => 0.0,
            "pna_end" => 0.0,
            "prc_end" => 0.0,
            "prd_end" => 0.0,
            "pra_end" => 0.0,
            "degree" => 0.0,
            "inclusion" => 0.0,
            "coopFreq" => 0.0,
            "fitness" => 0.0,
            "shortestPath" => 0.0,
            "connComponents" => 0.0,
            "meanConnCompSize" => 0.0,
            "largestConnComp" => 0.0,
            "meanConnDistance" => 0.0,
            "pxc_nc" => 0.0,
            "pxc_nd" => 0.0,
            "pxc_na" => 0.0,
            "pxc_rc" => 0.0,
            "pxc_rd" => 0.0,
            "pxc_ra" => 0.0,
            "pxd_nc" => 0.0,
            "pxd_nd" => 0.0,
            "pxd_na" => 0.0,
            "pxd_rc" => 0.0,
            "pxd_rd" => 0.0,
            "pxd_ra" => 0.0,
        ])
function hmap(data::String, range::Int, index::Int, titles::String, cLow::Float64, chI::Float64) 
    #indexes : 1 yidx   2 xidx   3 coopFreq   4 degree   5 fitness  6 pnc_end   7 pnd_end   8 prc_end  9 prd_end  
    #10 shortestPath 11 connComponents 12 meanCCSize 13 largestCC 14 mean connection distance 15 pna_end  16 pra_end  17 inclusion 
    #18 pc-nc 19 pc-nd 20 pc-na 21 pc-rc 22 pc-rd 23 pc-ra 24 pd-nc 25 pd-nd 26 pd-na 27 pd-rc 28 pd-rd 29 pd-ra
    dataArr = dataRead(data, range, index)
    x_axis = String[]
    y_axis = String[]
    for(i) in 1:range #PNC/PND times 10
        push!(x_axis, (string(round((0.4*i/range); digits = 4))))
        push!(y_axis, (string(round((10*i/range); digits = 3))))
    end
    p = heatmap(x_axis, y_axis,  dataArr[:, :, 1]; title = titles, xlabel="CL",ylabel="BEN", clim=(cLow,chI))
    p
end

#redefined hmap() in between runs

hmap (generic function with 3 methods)

file = "pc_d_pnd_fixed.csv"
fig1hmaps = [hmap(file, 20, 3, "Cooperation Frequency", 0.0, 1.0), 
hmap(file, 20, 6, "PNC", 0.0, 1.0), 
hmap(file, 20, 7, "PND", 0.0, 1.0),
hmap(file, 20, 8, "PR", 0.0, 0.8),
hmap(file, 20, 18, "P(C)NC", 0.0, 1.0),
hmap(file, 20, 19, "P(C)ND", 0.0, 1.0),
hmap(file, 20, 21, "P(C)R", 0.0, 1.0),
hmap(file, 20, 24, "P(D)NC", 0.0, 1.0), 
hmap(file, 20, 25, "P(D)ND", 0.0, 1.0), 
hmap(file, 20, 25, "P(D)R", 0.0, 1.0)] 

plot(fig1hmaps..., colorbar = true, grid=false, layout = (5,2), size = (1000, 1500))

Supplemental Figure 8 Probabilities among only cooperators and only defectors for PNC, PND, PNA, PR, and PRA in a run with PND evolving separately from PNC.